SENTINEL LYMPH NODE BIOPSY IN ORAL CAVITY SQUAMOUS CELL CARCINOMA WITHOUT CLINICALLY EVIDENT METASTASIS

Transcription

1 SENTINEL LYMPH NODE BIOPSY IN ORAL CAVITY SQUAMOUS CELL CARCINOMA WITHOUT CLINICALLY EVIDENT METASTASIS Risto Kontio, MD, 1 I. Leivo, MD, PhD, 2 E. Leppänen, MD, 3 T. Atula, MD, PhD 4 1 Department of Maxillofacial Surgery, Helsinki University Central Hospital, Kasarmikatu 11 13, Helsinki, Finland. risto.kontio@hus.fi 2 Department of Pathology, Helsinki University Central Hospital, Helsinki, Finland 3 Department of Laboratory Diagnostics and Nuclear Medicine, Helsinki University Central Hospital, Helsinki, Finland 4 Department of Otorhinolaryngology, Head and Neck Surgery, Helsinki University Central Hospital, Helsinki, Finland Accepted 5 May 2003 Published online 2 December 2003 in Wiley InterScience ( DOI: /hed Abstract: Background. The clinically N0 neck in patients with oral SCC is commonly treated by neck dissection because the existence of metastases cannot be excluded. To determine whether unnecessary treatment could be avoided, we evaluated the feasibility of sentinel lymph node (SLN) biopsy. Methods. Fifteen previously untreated patients with T1 or T2 oral SCC without clinically or radiologically detectable metastasis were included. A blue dye and gamma probe were used to identify SLNs. SLNs were stained with cytokeratins. All nodes in neck dissection specimens were stained using H & E. Results. SLNs were identified in 14 patients by lymphoscintigraphy and in all patients when probe and dye were combined. Four neck dissection specimens contained four metastatic lymph nodes. Three of the four lymph nodes were SLN. One SLN was found to be metastatic after immunostaining. However, although there was one blue sentinel node in one neck, a metastatic non-sln was present. Conclusions. Our results show that SLN biopsy is a promising tool for use in patients with oral SCC. However, further studies are necessary. A 2004 Wiley Periodicals, Inc. Head Neck 26: 16 21, 2004 Correspondence to: R. Kontio B 2004 Wiley Periodicals, Inc. Keywords: metastasis; neck dissection; oral cancer; sentinel lymph node Despite the quality of current imaging methods, the risk of occult metastasis in necks categorized as N0 in patients with squamous cell carcinoma (SCC) of the head and neck region is still between 20% and 30%. 1 In addition to being related to tumor site and size, the risk is related to depth of infiltration. 2 Treatment of the N0 neck is controversial. The main options are neck dissection, radiotherapy, a combination of the two, or simply a wait and see policy. 3 In many institutions, the neck is treated when risk of occult metastasis exceeds 20%. Most patients concerned are therefore over treated. Sentinel lymph node (SLN) biopsy has recently been used in connection with the treatment of oral and oropharyngeal SCC. If successful, SLN biopsy could allow neck dissection to be avoided in patients with no lymph node metastasis in the 16 Sentinel Lymph Node in Head and Neck Cancer HEAD & NECK January 2004

2 neck. SLN biopsy is based on the theory that small numbers of regional guardian lymph nodes collect metastatic cells. 4 Cabanas 5 was the first to use SLN biopsy more than 20 years ago during operations to treat penile malignancies. SLN biopsy has since become popular in the treatment of breast cancer and malignant melanoma. The sensitivity of SLN biopsy in connection with breast cancer and melanoma surgery has been estimated to be more than 95%. 6 The feasibility of SLN biopsy in head and neck SCC remains questionable. Results of only a few studies, involving limited numbers of patients, have been published Some of the studies included patients with clinically evident metastasis. Koch et al 8 pointed out that the technique is associated with various substantial problems. They were able to identify SLN in only two of five patients examined. Some metastatic lymph nodes were missed. Shoaib et al 9 used both radioactive tracer and a blue dye in 16 patients and found the method accurate in all seven necks containing impalpable metastases. In 2001, Shoaib et al 12 published a report on 40 cases with necks clinically classified as N0. Twenty contained subclinical metastases. SLNs were found in 17 of these necks. Sixteen contained metastases in SLNs. Alex et al 7 identified SLNs in all eight patients they studied, but only one patient had metastatic lymph nodes. Recently, Taylor et al 10 reported on nine patients, four of whom had evident metastasis. In these four patients, the SLNs were the only metastatic lymph nodes. The method has also been studied in head and neck melanoma by Jansen et al, 13 who reported that SLN biopsy is technically demanding in the head and neck region and does not seem to be as successful there as in other locations. They found 80% of the metastatic lymph nodes present. In this study, we assessed the feasibility of SLN biopsy in oral SCC. The study included patients with T1 or T2 carcinoma with no clinical or radiologic signs of neck metastases. MATERIAL AND METHODS Inclusion criteria were SCC in the oral cavity, tumor stage T1 or T2, no signs of metastasis as revealed by physical examination, MRI or CT, and neck dissection planned as part of a treatment strategy that involves neck dissection when primary tumors are larger than 1 cm and infiltration is deeper than 2 mm. Between December 2000 and August 2001, we enrolled 15 consecutive patients who had been referred to Helsinki University Central Hospital. Informed consent was obtained from all patients. The ethics committee approved the study protocol. Twenty hours before surgery, 0.2 ml of Nanocoll fluid ( 99m Tc-nanocolloid; Nycomed Amersham, High Wycombe, Bucks, UK) was injected peritumorally, anteriorly and posteriorly. The mean particle size in Nanocoll is less than 80 nm. The total emission of the volume injected is 74 MBq (2 mci). Static lymphoscintigraphy was performed 1 and 2 hours after injection. The following day, immediately before neck dissection, 2.0 ml of Patent Blue dye (Laboratoire Guerbet, Aulnay-Sous-Bois, France) was injected into the same sites as those into which Nanocoll had been injected. During dissection, a Neo2000 Gamma Detection System (Neoprobe Corp, Columbus, OH) was used to identify radiolabeled nodes. A handheld gamma probe (Neoprobe) was equipped with a 14-mm probe with an external 15j angle collimator. A 20% window centered on the 140-keV photon peak was used. Target counts (2 to 6 seconds) and target counts vs background counts were measured both from the primary tumor and from radiopositive (hot) nodes. Neck dissection, at levels I III or I IV, was carried out in all patients. One patient with a midline tumor underwent bilateral neck dissection, resulting in dissection of 16 neck sites. During neck dissection, SLNs (hot, hot and blue, or cold but blue) were identified and removed separately. Six-micrometer sections were cut from both SLNs and other lymph nodes at intervals of not more than 5 mm. If hematoxylin-eosin staining revealed no metastatic growth, immunohistochemistry was performed on the SLN. The primary antibodies used were cytokeratin CAM 5.2, high-molecular weight cytokeratin, and cytokeratin AE1/AE3. RESULTS Of the 15 consecutive patients, eight were men and seven were women. The mean age of the patients was 63.8 years (range, years). The clinical data are summarized in Table 1. Lymphoscintigraphy revealed SLNs in 14 of the 15 patients. Thirty-seven radiopositive lymph nodes, regarded as SLNs, were observed (Table 1). In the first patient studied, there were no radiopositive SLNs, but there were Patent Blue Sentinel Lymph Node in Head and Neck Cancer HEAD & NECK January

3 Table 1. Clinical data relating to the 15 patients in whom sentinel lymph nodes were examined. Age/ Sex Site of tumor pt Neck surgery Ipsilateral LSC/SN/ML Contralateral LSC/SN/ML Ipsilateral H/HB/B Contralateral H/HB/B 69/M FOM 2 L1 L3* 0/1/1 0/1/1 0/0/1 0/0/1 80/F Tongue 2 L1 L3 2/3/0 1/1/1 47/M FOM 1 L1 L4 y 2/3/1 1/1/1 0/1/0 72/F Tongue 2 L1 L3 y 1/2/0 1/1/0 1/1/0 0/2/0 35/F Tongue 2 L1 L4 2/3/0 2/2/0 1/1/1 74/F FOM 1 L1 L3 2/1/1 2/0/0 0/1/0 50/M FOM 2 L1 L3 1/1/0 0/1/0 81/F Buccal 2 L1 L3 2/4/0 1/1/2 mucosa 56/F Tongue z 1 L1 L4 3/7/0 2/0/0 2/5/0 78/M FOM 1 L1 L4 2/4/0 2/2/0 66/M Tongue 1 L1 L3 y 3/4/0 2/1/0 1/2/1 1/0/0 65/M Tongue 1 L1 L3 3/3/0 3/0/0 3/0/0 49/M FOM 1 L1 L4 2/2/0 0/2/0 60/F Tongue 1 L1 L3 2/3/0 0/3/0 47/M Retromolar 2 L1 L4 2/1/0 0/1/0 Abbreviations: LSC, numbers of SLN detected by preoperative lymphoscintigraphy; SN, numbers of SLN detected intraoperatively using a gamma probe or Patent Blue dye; ML, numbers of histopathologically metastatic lymph nodes. SLN recorded as hot (H), hot and blue (HB), or blue (B) in two right-hand columns; FOM, floor of mouth. *Bilateral neck dissection. y SLN biopsy also on contralateral side. z Basaloid squamous cell carcinoma (all others are SCC). stained nodes. Six patients also had contralateral SLNs nodes (Figure 1.) All ipsilateral SLNs detected by means of scintigraphy were identified during surgery. Of 12 contralateral scintigraphically positive nodes (in six patients), only four (in three patients) were identified and removed for histopathologic examination. The rest were monitored clinically and by CT examination. The median follow-up time of these patients with no signs of metastases was 8.8 months (mean, 9.3 months). During surgery, 43 SLNs (14 hot, 21 hot and blue, and 8 cold but blue) (Table 1) were identified and sent for histologic analysis. The mean number of lymph nodes from neck dissection specimens per patient was 13, including SLNs (range, 5 29). Hematoxylin-eosin staining revealed three metastatic nodes in three neck sides (two patients). Two of these lymph nodes were SLNs. Immunohistochemical staining of SLNs revealed one further metastatic lymph node (in the first patient) (Figure 2). Three of the 15 patients (20%) (four neck sides), therefore, had metastatic lymph nodes (Table 2). FIGURE 1. One contralateral lymph node detected in lymphoscintigraphy. Primary tumor (floor of mouth) cranially and ipsilaterally with two sentinel lymph nodes visible. FIGURE 2. In one patient, only cytokeratin immunohistochemical staining revealed metastatic growth in harvested sentinel lymph nodes. 18 Sentinel Lymph Node in Head and Neck Cancer HEAD & NECK January 2004

4 Table 2. Total number of SLNs found perioperatively. Shaded squares imply true metastatic nodes. Note one contralateral non- SLN metastasis with nonmetastatic SLN. Ipsilateral neck No. H H&B B Abbreviations: H, hot; H&B, hot and blue; B, blue. In two patients who had metastatic lymph nodes, the SLN was the only one that contained tumor growth. In the remaining patient, who underwent bilateral neck dissection, one SLN (blue) was identified on each side. One contained metastatic growth; the other did not. However, one additional metastatic lymph node was found on hematoxylin-eosin staining of material from the neck side on which the negative SLN was found. DISCUSSION Metastatic patterns in head and neck SCC are to a certain extent predictable. 14 However, as with other solid carcinomas, lymphatic flow is unpredictable in individual cases. 4 Identification of SLNs in the neck would facilitate examination of critical lymph nodes. The principal idea behind SLN biopsy is prevention of neck dissection in cases in which SLNs are free from disease. Patients who would benefit from the method are, therefore, those with no clinically evident metastasis in the neck. Technically, the method is particularly suitable in cases of oral or oropharyngeal tumors, but it has also been used in cases of laryngeal tumors. 7 The incidence of occult metastasis in our study was 20% (three of 15 patients), in line with generally reported incidences. Three of four metastatic lymph nodes in four neck sites were SLNs. A major concern, however, is that one metastatic lymph node was not an SLN. The patient concerned was the only patient in whom no SLN was detectable on lymphoscintigraphy. Our results are similar to those of Jansen et al. 13 They found SLNs on lymphoscintigraphy in 90% of patients. In breast cancer, 2% to 7% of patients have been found to have no SLNs on lymphoscintigraphy. In such cases, surgery in relation to all lymph nodes at the risk site is recommended. 6 One reason why no radiopositive node is seen on lymphoscintigraphy might be that a metastatic lymph node has ceased to be functional, and no colloid can, therefore, concentrate in it. It has been suggested that replacement of a lymph node by a bulky metastatic cancer can result in diversion of lymphatic flow and radiolabeling of downstream nodes. 8 This suggestion has been supported by Krag et al. 6 Another reason for a metastasis being missed might be that a radioactive SLN is too close to the primary tumor and, therefore, remains undetected. 8 Shoaib et al 9 suggested this as a possibility in cases in which no SLNs are found on lymphoscintigraphy. Alex et al 7 suggested that the problem might be solved by changing the angulation of the handpiece, shielding the primary tumor with lead, and adjusting the threshold parameters of the scintillation counter. We had one false-negative SLN in our study. The lymphatic drainage system is known to be complex in the head and neck region. The slight difference of site and depth of injection might cause the radiocolloid or the dye to drain along different routes, affecting which lymph nodes become positive. The injected fluid does not necessarily drain to a metastatic lymph node as occurred in our case. We carried out neck dissection before removing the primary tumor, because we wanted to retrieve blue lymph nodes before color had drained away. Despite doing so, we found no significant difficulties arising from radioactivity of the primary tumor. Taylor et al 10 have, however, stated that initial removal of the primary tumor is essential. We used a narrow-angle (15j) collimator. Use of Sentinel Lymph Node in Head and Neck Cancer HEAD & NECK January

5 such a collimator would seem essential if a primary tumor is not initially removed. However, use of a wider-angle collimator might result in more rapid identification of SLNs. Resecting the primary tumor first might have helped in the search of hot nodes, but on the other hand, we observed that the blue dye spread rapidly, resulting in multiplying SLNs. Lead shields have also been used to protect the emission from the primary tumor. Lymphoscintigraphy revealed 25 ipsilateral and 12 contralateral lymph nodes. At surgery, all ipsilateral SLN were found, but four were not detected until the neck specimen was removed and examined on the operating table. Our findings are similar to those of Shoaib et al. 9 They identified two of 32 SLNs after completion of neck dissection. Jansen et al 13 were unable to find 15% of SLNs identified by lymphoscintigraphy. At surgery, radioactivity of the primary tumor and SLN was found to vary widely. The fact that more SLNs were identified during neck dissection than by lymphoscintigraphy was a consequence of including in the total SLN count lymph nodes that were not radiopositive but contained dye. All identified SLNs were not both hot and blue. Most SLNs were, but we had both plain hot and plain blue nodes. Plain blue nodes, however, were in the minority. At least two reasons might explain the difference. The slight difference of the site of injection might reflect where the colloid or dye flows. Another explanation might be the difference of the speed of the flow between the Nanocoll and blue dye. Lack of clarity of a lymphoscintigraphic image presents a problem, as Jansen et al 13 also reported. Detailed lymphoscintigraphic images help in locating SLNs. Spillage of isotope in the pharynx and esophagus can cause problems. 8 We were able to avoid such difficulties by asking patients to wash their mouths out immediately after injection, as recommended by Shoaib et al. 9 Contralateral SLNs seen on lymphoscintigraphy presented an ethical problem, because they were not at a high-risk site and would not have been removed had our earlier surgical protocol been followed. We removed six radiopositive contralateral nodes, which were easily located with the radioprobe. The rest were simply monitored. Because routes of metastasis can be unpredictable, biopsy of contralateral SLN might be warranted. By use of Nanocoll, we found on average 1.7 ipsilateral SLNs (range, 0 3) and 0.8 contralateral SLN (range, 0 3) on lymphoscintigraphy. Four patients had four or more SLNs. On the basis of SLN theory, there should be no more than one to two SLNs. 4 In the neck region, the anatomy of which is complex, removal of all observed SLNs made operations longer and more difficult than would otherwise have been the case. One risk of SLNs biopsy might be tumor spread, if metastatic SLN are removed individually. Whether the procedure, in fact, enhances tumor spread is not yet known. Jansen et al, 13 who also used Nanocoll, reported that identification of large numbers of lymph nodes by means of lymphoscintigraphy caused problems. Because the particle size of Nanocoll is small, it might cause the fluid to spill through the SLNs, and a larger number of lymph nodes will become positive. Larger particle size should be tried as suggested by Jansen et al. 13 An alternative fluid is Albures, which was used by Shoaib et al. 9 The optimal particle size in relation to the head and neck region remains to be determined. The scintigraphic agent commercially available in the United States is sulfur colloid with a larger particle size than either Nanocoll or Albures. Because the particle size intrinsically is the most important characteristics of the scintigraphic agent, the use of sulfur colloid products should produce similar results with possibly somewhat fewer SLNs detected. This topic, however, remains to be resolved. It is questionable whether the use of any dye is meaningful, because the principal aim of the method is to retrieve only SLNs by means of a limited neck incision. However, we found the use of a combination of dye and radioactive tracer valuable. Stained lymphatic pathways are visible, allowing them to be easily followed and facilitating location of SLNs. Pitman et al 15 used a blue dye without radioactive tracer to label lymph nodes in cases of SCC of the head and neck. However, they terminated their study after 16 patients, because they were unable to identify any SLNs. Shoaib et al 9 used only dye initially. They found the method unsuccessful. Later, using both radioactive tracer and a blue dye, they found SLNs in 15 of 16 patients. Jansen et al 13 also suggested the use of both a dye and a radioactive tracer. Some investigators have used radioactive tracer without dye. Koch et al 8 experienced difficulty in identifying SLNs in five patients, but Alex et al 7 identified SLNs in all eight of their patients. The injections of blue dye changed the color of oral tissues, but this did not cause difficulties in 20 Sentinel Lymph Node in Head and Neck Cancer HEAD & NECK January 2004

6 evaluation of resection margins. One concern is the potential allergy. Allergic reactions to Nanocolloid during lymphoscintigraphy have been published. 16 Anaphylaxis to the blue dye has also been documented. 17 Although in our series no allergic reactions were seen, one must be aware of the potential serious risks of anaphylaxis. It is known that not all metastases might be found during routine histopathologic examination. 11 In our study, cytokeratin staining revealed one additional metastasis. Of the three different cytokeratins, AE1/AE3 proved to be the most sensitive one. It is not yet known how meticulously SLNs should be examined. We did not record how long the SLN biopsy took. It was performed during the neck dissection incision. Although the procedure is technically demanding, we found it relatively easy to learn. Removing SLNs requires substantial experience in neck surgery and extensive knowledge of metastatic spread patterns. We did not include patients with small, superficial T1 tumors, who are at low risk of metastasis. 18 With large tumors, the risk of metastasis is significantly greater, and neck dissection could not be replaced by SLN biopsy. If it proves successful, the method would seem most likely to be suitable for patients with intermediate-sized tumors (T2). Although one metastatic lymph node was not an SLN, our results encourage us to study the method further. Results of studies involving many more patients are needed before a decision can be reached as to whether the method might replace neck dissection. REFERENCES 1. Woolgar JA. Pathology of the N0 neck. Br J Oral Maxillofac Surg 1999;37: Moore C, Kuhns JG, Greenberg RA. Thickness as prognostic aid in upper aerodigestive tract cancer. Arch Surg 1986; 121: Pillsbury HC, Clark M. A rationale for therapy of the N0 neck. Laryngoscope 1997;107: Leong SPL. The role of sentinel lymph nodes in human solid cancer. Principles Pract Oncol 1998;4: Cabanas RM. An approach for treatment of penile carcinoma. Cancer 1977;39: Krag D, Weaver D, Ashikaga T, et al. The sentinel node in breast cancer a multicenter validation study. N Engl J Med 1998;339: Alex JC, Sasaki CT, Krag DN, Wenig B, Pyle PB. Sentinel lymph node radiolocalization in head and neck squamous cell carcinoma. Laryngoscope 2000;110: Koch WM, Choti MA, Civelek AC, Eisele DW, Saunders JR. Gamma probe-directed biopsy of the sentinel node in oral squamous cell carcinoma. Arch Otolaryngol Head Neck Surg 1998;124: Shoaib T, Soutar DS, Prosser JE, et al. A suggested method for sentinel node biopsy in squamous cell carcinoma of the head and neck. Head Neck 1999;21: Taylor RJ, Wahl RL, Sharma PK, et al. Sentinel node localization in oral cavity and oropharynx squamous cell cancer. Arch Otolaryngol Head Neck Surg 2001;127: Stoeckli SJ, Pfaltz M, Steinert H, Schmid S. Histopathological features of occult metastasis detected by sentinel lymph node biopsy in oral and oropharyngeal squamous cell carcinoma. Laryngoscope 2002;112: Shoaib T, Soutar DS, MacDonald DG, et al. The accuracy of head and neck carcinoma sentinel lymph node biopsy in clinical N0 neck. Cancer 2001;9: Jansen L, Kroops HS, Nieweg OE, et al. Sentinel node biopsy for melanoma in the head and neck region. Head Neck 2000;22: Byers RM, Weber RS, Andrews T, McGill D, Kare R, Wolf P. Frequency and therapeutic implications of skip metastases in the neck from squamous cell carcinoma of the oral tongue. Head Neck 1997;19: Pitman KT, Johnson J, Edington H, et al. Lymphatic mapping with isosulfan blue dye in squamous cell carcinoma of the head and neck. Arch Otolaryngol Head Neck Surg 1998;124: Burton DA, Cashman JN. Allergic reaction to nanocolloid during lymphoscintigraphy for sentinel lymph node biopsy. Br J Anaesth 2003;90: Cashman JN. Editorial Sentinel lymph node biopsy: Anaesthetic implications. Eur J Anaesthesiol 2001;18: Persky MS, Lagmay VM. Treatment of the clinically negative neck in oral squamous cell carcinoma. Laryngoscope 1999;109: Sentinel Lymph Node in Head and Neck Cancer HEAD & NECK January